Performance Analysis in Volleyball: Problem of Merging Attack and

Counterattack Spike as One Variable

Ivana Klaričić, Hrvoje Ajman and Josip Cvenić

Faculty of Kinesiology, University of Osijek, Osijek, Croatia

Keywords: Volleyball, Situational Efficiency, Game Phases, Attack Complex, Counterattack Complex.

Abstract: The purpose of this study was to present the multiple regression model with attack and counterattack spike as

separate variables is more appropriate than the model with the attack and counterattack spike merged as one

variable. Two multiple regression analyses were conducted that determine the relationship between situational

efficiency parameters of volleyball game phases with the set score. Game phases included into both regression

models were the serve, reception, spike, block and dig. One of the regression models had an attack and

counterattack spike as two separate variables and the other one had them merged as one. A sample was 40

randomly selected volleyball sets played in the European League for Men in 2011 and 2012. Although the

sample wasn't recent, the purpose of this methodologically based study was to present deficiency of merging

attack and counterattack spike as one variable. Both multiple regression analyses determined a high and

positive relationship between the situational efficiency of volleyball game phases with the set score. The spike

as a merged variable had 32.5% of common variance with the set score. But when separated into two variables,

the attack spike had only 8.9% of common variance and counterattack spike 26.5%. Although the spike was

the game phase that had the highest relationship with the set score, the spike in the counterattack was the one

that contributed. Ultimately the serve, reception and dig had higher common variance with the set score then

the attack spike. Conclusion was that the attack and counterattack spike need to be considered as separate

variables because of specificity of the attack and counterattack complexes of the volleyball game.

1 INTRODUCTION

The volleyball game consists of six phases that are

sequentially executed, the serve, reception, setting,

attack, block and dig (Busca and Febrer, 2012). The

game phases that one team sequentially executes are

organised as a game complex. The two main game

complexes in volleyball are an attack and

counterattack. The attack complex consists the

reception, setting, attack spike and the counterattack

the serve, block, dig, setting, counterattack spike.

Both the attack and counterattack complex have their

specificities and should be considered separately. The

attack has more predictable conditions and, more

importantly, a structured attack that takes place in

specific sequences. On the other hand, the

counterattack is characterized by a less structured,

slower game that is the result of more variable

conditions in which the counterattack begins

(Marcelino, et al., 2009). After the attack spike, the

ball is faster, has a straight trajectory and involves the

participation and interconnection of a greater number

of game factors (Afonso, et al., 2005).

Among all volleyball phases, points are mostly

achieved by successful spikes (Marcelino, et al, 2008)

followed by blocks and serves and opponent's

mistakes. Spike is the phase of the volleyball game

that shows the highest correlation with winning

(Marcelino, and Mesquita, 2006). In top volleyball,

the average number of points scored in a match is 45.5

by spike, 10.0 by block and 5.0 by serve (Marcelino,

and Mesquita, 2006). Furthermore, Barzouka, et al.

(2006) also determined spike's relationship with the

score, but both in the attack and counterattack. The

frequency of strong spikes is significantly higher in

the attack phase (Castro, and Mesquita, 2008), with

an emphasis on a faster pace of game in the same

phase (Afonso, et al., 2005). In contrast, in the

counterattack, the tempo of the attack in the game is

slower, which reduces the probability of winning a

point, allowing the opponent's block to have more

blockers (Afonso, et al. 2005).

The most important characteristic of volleyball is

that the game phases are executed in a manner that an

efficiency of every game phase is partially

determined by the previous one. According to the

Klari

c, I., Ajman, H. and Cveni

c, J.

Performance Analysis in Volleyball: Problem of Merging Attack and Counterattack Spike as One Variable.

DOI: 10.5220/0012939200003828

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 12th International Conference on Sport Sciences Research and Technology Support (icSPORTS 2024), pages 54-59

ISBN: 978-989-758-719-1; ISSN: 2184-3201

aforementioned sequentiality of volleyball game, the

phases that the team can not win the point are not

irrelevant for the score. The preceeding phases impact

spike's efficiency in a different manner in the attack

and counterattack. Costa, et al. (2010) determined

that a stronger serve greatly reduces the quality of the

opponent's attacking actions. Papadimitriou, et al.

(2004) determined that the quality of reception

significantly differentiates the attack tactics chosen

by the setter. Barzouka, et al. (2006) also determined

that the excellent performance of the Olympic-level

setters and spikers is highly related to the

performance of the actions that preceded them. The

results showed that the frequency of the setter's

excellent performances is significantly higher when

they are preceded by an excellent reception compared

to a good reception (49.0 % : 23.4 %). Similarly,

spikers had a higher frequency of excellent

performance when the setter had an excellent set,

compared to a very good or only good one (79.4 % :

51.4 %; 79.4 % : 28.3 %) (Barzouka, et al., 2006).

Some reserchers considered spike as a merged

variable in their performance analysis studies

(Yiannis and Panagiotis, 2005; Valladares, et al.,

2016; Silva, et al., 2014). Also, some researchers

separated variables related to different complexes

(Stutzig, et al., 2015). Given the specificities of the

attack and counterattack complex in volleyball, the

assumption was that attack and counterattack spike

would have different relationship with the set score.

The purpose of this study is to present the multiple

regression model with attack and counterattack spike

as separate variables is more appropriate than the

model with attack and counterattack spike merged as

one variable.

2 METHODS

The data were collected from the existing videos of

volleyball matches. It was done by the first author,

who has multiannual playing experience, an A

coaching license and a multiannual coaching

experience in men’s volleyball. The reliability

analysis was conducted with the help of an expert

with multiannual playing, coaching and notational

analysis work experience.

2.1 Set of Entities

The sample was 40 volleyball sets from matches

played in the European Volleyball League for Men in

2011 and 2012. Only the data from one set of a match

and only one team were collected in order to avoid

interdependence of the sets. Both the team and the set

were randomly selected.

2.2 Set of Variables

The predictor variables were the efficiency

coefficients of the five phases of the volleyball game:

serve, reception, spike, block, dig, and their intrateam

variability. The setting was excluded from this study

because of its specific situational efficiency analysis.

The efficiency coefficient of each game phase was

defined as the arithmetic mean of scores of all

performed technical skills within a particular phase in

one set. Each performed skill was evaluated with a

score (1 – 4) according to precisely defined criterion.

The score 1 was an error, 2 was an advantage for the

opponent, 3 was an advantage for the team being

evaluated, and 4 was an ideal performance (reception,

dig) or a point won (serve, spike, block). The criterion

variable was the set score defined as a relative point

difference, the point difference in a set divided by the

total number of points. If the team won the set, the

relative point difference was positive and on the

contrary, if the team lost the set, the relative point

difference was negative. The authors believe that the

same point difference does not represent an equal

outcome of the set in the case when the result is 15 :

13, 25 : 23 or 31 : 29. For this reason, the result in the

set was defined as a relative point difference.

2.3 Statistical Analysis

A reliability analysis was conducted on a sample of 3

randomly selected sets. Spearman’s rank correlation

and Cohen's kappa were calculated to determine the

degree of agreement between the two different

measurements of the same measurer (the first author)

and two different measurements (the first author and

the expert) at intervals of 4 – 6 weeks (test-retest

method).

The descriptive statistics were: arithmetic mean

(Mean), standard deviation (σ), minimum (Min) and

maximum (Max). Normality of distribution was

determined by the Shapiro-Wilk test.

Two separate multiple regression analyses were

conducted to determine the relationship between the

efficiency coefficients of the volleyball game phases

and the relative point difference in the set. The first

one was conducted with the attack and the

counterattack spike merged as one variable. The

second one was conducted with attack and the

counterattack spike as two separate variables.

Performance Analysis in Volleyball: Problem of Merging Attack and Counterattack Spike as One Variable

The collected data were analysed with the

computer program Statistica for Windows 13.3

(TIBCO Software Inc.).

3 RESULTS

Reliability analysis results determined a high

correlation between the two measurements of the

same measurer conducted at two-time points (R =

0.91; κ = 0.92) and the two different measurers (R =

0.92; κ = 0.88).

Two separate multiple regression analyses were

conducted in order to determine the relationship of

the efficiency coefficients of volleyball game phases

and the relative point difference in the set. The first

model had attack and counterattack spikes merged as

one variable. The second model had them as two

separate variables with a total of six game phases.

Both regression analyses showed that all

predictors had a high and significant relationship with

the set score. The efficiency coefficients of phases of

the volleyball game and the variability of the phases

explained a total of 80.0 and 80.7% of the variance of

the relative point difference in the set. The first

multiple regression analysis showed that spike was a

game phase with the highest amount of common

variance with the score, 32.6%. But when spike was

separated into attack and counterattack spike, attack

spike had 8.8% of common variance with score whilst

the counterattack spike had 26.5 which is 3 times

higher.

Table 1: Descriptive statistics results.

Mean ± σ Min Max

Relative point difference

-0.01 ± 0.13 -0.32 0.25

Efficiency coefficient – serve

2.14 ± 0.20 1.75 2.50

Efficiency coefficient – reception

2.98 ± 0.26 2.55 3.50

Efficiency coefficient – spike

3.04 ± 0.24 2.44 3.65

Efficiency coefficient – att. spike

3.12 ± 0.29 2.40 3.82

Efficiency coeff. – counteratt. spike

2.95 ± 0.40 1.75 4.00

Efficiency coefficient – block

2.29 ± 0.39

1.00 3.00

Efficiency coefficient – dig

1.95 ± 0.28

1.22 2.61

Legend: Mean – arithmetic mean, σ – standard deviation,

Min – minimal result, Max – maksimal result.

Table 2: The results of two multiple regression analyses (attack and counterattack spike as two separate variables in the second

analysis).

β b t

part. (%)

R 0.89 Intercept

-2.29 -11.31

0.00

80.0% Efficiency coefficient-serve

0.35 0.25 4.37

16.7

0.00

adj

76.5% Efficiency coefficient-reception

0.27 0.14 3.31

11.9

0.00

26.3 Efficiency coefficient-spike

0.48 0.26 5.74

32.5

0.00

p 0.00 Efficiency coefficient-block

0.22 0.08 2.65

5.3

0.01

Efficiency coefficient-dig

0.39 0.19 4.69

12.9

0.00

R 0.90 Intercept

-2.17 -10.64 0.00

80.7% Efficiency coefficient-serve

0.36 0.25 4.45

17.1

0.00

adj

77.2% Efficiency coefficient-reception

0.29 0.15 3.63

12.8

0.00

23.0 Efficiency coefficient-att. spike

0.20 0.10 2.33

8.9

0.03

p 0.00 Efficiency coefficient-counteratt. spike

0.42 0.14 4.70

26.5

0.02

Efficiency coefficient-block

0.15 0.05 1.70

3.6

0.10

Efficiency coefficient-dig

0.37 0.18 4.48

12.2

0.00

Legend: R – coefficient of multiple correlation, R

– coefficient of determination,

adj

– adjusted coefficient of

determination, F – Fisher's test value, β – standardized regression coefficients, b – unstandardized regression coefficients,

part

– partial coefficient of determination,

t – t–test value, p – significance level.

icSPORTS 2024 - 12th International Conference on Sport Sciences Research and Technology Support

4 DISCUSSION

The purpose of this research was to present the

multiple regression model with the attack and

counterattack spike as separate variables as a more

appropriate model than the one with the attack and

counterattack spike merged as one variable.

Descriptive parameters showed that the attack

spike is the game phase that had the highest

situational efficiency coefficient, 3.12 out of 4, the

maximal possible efficiency coefficient. The

counterattack spike had a lower efficiency coefficient

than the attack spike, 2.95. The game phase with the

second highest efficiency coefficient was the

reception with almost equal values as the

counterattack spike. Next one was the block and

serve, and the dig had the lowest situational efficiency

coefficient, 1.95. The counterattack spike also had a

higher standard deviation then the attack spike (0.40),

the highest standard deviation of all game phases. It

means that the teams were the most heterogeneous in

the counterattack spike. This could be due to unstable

conditions where the counterattack spike is

performed. For comparison, serve had the lowest

standard deviation, 0.20. The reason is that serve has

the most stable conditions for execution, with no

preceding game phase to disturb the performance.

However, a situational efficiency coefficient of a

game phase does not represent its relationship with

the set score. So multiple regression analyses were

conducted to determine the aforementioned

relationship. The results of both multiple regression

analysis showed that the situational efficiency

coefficient of the game phases had a high and

significant relationship with the set score, 80.0% and

80.7%. In the first model, all regression coefficients

of game phases were positive, an increase in their

efficiency coefficients had a positive impact on the

set score. The spike was the game phase that

explained the most variance of the set score (32.5%),

followed by serve (16.7%), dig (12.9%) and reception

(11.9%) and finally block with only 5.3%.

Unlike the first regression model, the model with

the attack and counterattack spike separated, block

had no significant relationship with the score. Even in

the first multiple regression model the block had the

lowest common variance with the score. So when the

spike was separated into two variables, the

counterattack spike took over some of the block's

common variance with the score due to its significant

intercorrelation (0.40) and the block remaind with

unsignificant relationship with the score. Statistically

insignificant relationship of the block with the score

could be unexpected. Marcelino et al. (2008)

determined that the block points were a high indicator

for success in volleyball. But they considered only the

phases that win the points, the serve, spike and block,

and each phase separately. Due to phases'

intercorrelations and the ones with the other game

phases, multiple regression parameters for the block

are expected to be lower.

Also the attack spike and the counterattack spike

had very different amounts of common variance with

the score in the second regression model. The attack

spike had 8.9% of common variance with the score

and the counterattack spike 26.5% which is 3 times

higher. As mentioned, a high efficiency coefficient

does not imply a high impact on the set score. So

attack spike being the game phase with the highest

efficiency coefficient of all phases, had the second

lowest amount of common variance with the set

score. Contrary to the attack spike, counterattack

spike had 26.5% of common variance with the score.

It means that counterattack spike is the game phase

that differentiates a winning from a losing set.

Stutzig, et al. (2015) also determined that the best

predictors for the score and the team level are the

variables related to complex 2 (effectivity of counter-

attack, effectivity of medium and slow attack-tempo)

whilst the impact of complex 1 variables (action

sequences of reception, setting and attacking) were

marginal. Drikos, et al. (2021) state in their study that

K1 (attack complex) does not differentiate teams of

various performance levels. Even the weakest teams

in a tournament can achieve a high success rate by

playing under ideal conditions in K1. On the contrary,

the variable differentiating the performance level

between teams ranked in the upper group and the two

other groups of lower-ranking was the effectiveness

of attack after the reception (Drikos, et al., 2021).

The results also show that due to sequentiality of

the volleyball game, the phases that the team can not

win the point are not less relevant. The reception and

dig had an unexpectedly high amount of common

variance with the score considering that they are the

phases that the team can not win the points with. They

had a total of approximately 25% of common

variance with the set score.

Laporta, et al. (2017) state that there are even six

types of complexes in volleyball. Authors consider

the serve as a separate complex (K0). Then they

differentiate 5 complexes depending on the action the

complex begins with, the reception, block-dig

(serving team), block-dig (receiving team), attack

coverage, freeball/downball. They also state that

authors shouldn't incorporate K0, K3, K4 and K5 into

K2 (counterattack). This shows that after the attack

spike (K1) conditions of the game become more and

Performance Analysis in Volleyball: Problem of Merging Attack and Counterattack Spike as One Variable

more unstructured and with more possible variants.

Hileno, et al. (2020) introduced also undefined

complex (KU) referring to the actions that are

difficult to classify. But the regression analysis model

needs to be as simple as possible so including 5 or

even 6 different types of spikes according to the

complex would lead to other problems. The model

might become too difficult to explain.

Practical application in the training process would be

that the coaches have to produce the unstable

conditions that the counterattack spike is executed in.

Specifically, unstable conditions are reffered to as

many as possible variants of the counterattack

complex. Situational drills would be the better choice

then the competitive conditions like the volleyball

game itself. The drills that produce game situations

give coaches better control of the game factors being

practised. Practising in controlled conditions, the

team has the possibility of more repetitions of the

same game situation. Training in competitive

conditions like the volleyball game itself the same

game situation occurs many times less. The

counterattack spike is a game phase that is executed

mostly eight in a sequence and in unpredictable

conditions. So it would be too exhausting for the

players to accomplish necessary repetitions. But to be

noted, practising in the competitive conditions is the

best option when unpredictable situations have to be

produced.

5 LIMITATIONS AND

IMPLICATIOS

European League for Men is a top level volleyball

competition so the limitation of this study is that its

results could not refer to women's competition or

other levels of competition. It is difficult to assume

that the differences in the relationship with the set

score between attack and counterattack spike would

be similar if the volleyball sets were played in a

women's competition or lower level. It is possible that

in a women's competition, the attack spike is a phase

that better differentiate winning from losing sets

because spikes are not as strong as in men's

competition and the teams might be less

homogeneous. So the implication of this study is that

further research should be conducted with volleyball

sets collected from the lower level of competition or

from the women's competition.

6 CONCLUSIONS

The multiple regression model with the attack and the

counterattack spike as two separate variables showed

that counterattack spike had 3 times higher

relationship with the score. Also even three game

phases had a higher relationship with the score, the

serve, reception and dig. Only the block had a lower

relationship than the attack spike. The scientific

application of the results is to separate spike as the

attack and counterattack spike when possible due to

their specificity. The practical application of the

results of this research is a recommendation for teams

to place additional emphasis in the training process

primarily on increasing the efficiency of the

counterattack spike and also the phases that improves

its execution, the block and the dig. Training process

has to be more creative to produce the unstable

conditions that the counterattack spike is executed in.

REFERENCES

Afonso, J., Mesquita, I., & Palao, J. (2005). Relationship

between the tempo and zone of spike and the number of

blockers against the hitters. International Journal of

Volleyball Research, 8(1), 19 – 23.

Barzouka, K., Nikolaidou, M., Malousaris, G., & Bergeles,

N. (2006). Performance excellence of male setters and

attackers in Complex I and II on Volleyball teams in the

2004 Olympic Games. International Journal of

Volleyball Research, 99(1), 19 – 24.

Busca, B., & Febrer, J. (2012). Temporal fight between the

middle blocker and the setter in high level volleyball.

International Journal of Medicine and Science of

Physical Activity and Sport, 12(46), 313 – 327.

Castro, J., & Mesquita, I. (2010). Analysis of the attack

tempo determinants in volleyball's complex II - a study

on elite male teams. International Journal of

Performance Analysis in Sport, 10(3), 197 – 206.

Costa, G.C.T, Mesquita, I., Greco, P.J., Ferreira, N.N., &

Moraes, J. C. (2010). Relationship between tempo, type

and effect of attack in male young volleyball players

from high competitive level. Revista brasileira de

Cineantropometria e Desempenho Humano, 12(6), 428

– 434.

Drikos, S., Barzouka, K., Nikolaidou, M. E., &

Sotiropoulos, K. (2021). Game variables that predict

success and performance level in elite men’s volleyball.

International Journal of Performance Analysis in

Sport, 21(5), 767-779.

Hileno, R., Arasanz, M., & García-de-Alcaraz, A. (2020).

The sequencing of game complexes in women’s

volleyball. Frontiers in Psychology, 11, 529409.

Laporta, L., Afonso, J., & Mesquita, I. (2018). Interaction

network analysis of the six game complexes in high-

icSPORTS 2024 - 12th International Conference on Sport Sciences Research and Technology Support

level volleyball through the use of Eigenvector

Centrality. PloS one, 13(9), e0203348.

Marcelino, R., Cesar, B., Afonso, J., & Mesquita, I. (2009).

Attack-tempo and attack-type as predictors of attack

point made by opposite players in volleyball.

International Journal of Performance Analysis in

Sport, 9(3), 391 – 400.

Marcelino, R., & Mesquita, I. (2006). Characterizing the

efficacy of skills in high performance competitive

volleyball. In H. Dancs, M. Hughes & P. O’Donoghue

(Eds.), Notational Analysis of Sport - VII (pp. 491 –

496). Cardiff: UWIC.

Marcelino, R., Mesquita, I., & Afonso, J. (2008). The

weight of terminal actions in volleyball. Contributions

of the spike, serve and block for the teams' rankings in

the World League 2005. International Journal of

Performance Analysis In Sport, 88(2), 1 – 7.

Papadimitriou, K., Pashali, E., Sermaki, I., Mellas, S., &

Papas, M. (2004). The effect of the opponents serve on

the offensive actions of greek setters in volleyball

games. International Journal of Performance Analysis

in Sport, 4(1), 23 – 33.

Silva, M., Lacerda, D., & Joao, P. V. (2014). Game-Related

volleyball skills that influence victory. Journal of

Human Kinetics, 41(1), 173-179.

Stutzig, N., Zimmermann, B., Busch, D., & Siebert, T.

(2015). Analysis of game variables to predict scoring

and performance levels in elite men’s

volleyball. International Journal of Performance

Analysis in Sport, 15(3), 816-829.

Valladares, N., García-Tormo, J. V. & Joao, P.V. (2016)

Analysis of variables affecting performance in senior

female volleyball WorldChampionship 2014,

International Journal of Performance Analysis in

Sport, 16(1), 401-410.

Yiannis, L. & Panagiotis, K. (2005) Evolution in men’s

volleyball skills and tactics as evidenced in the Athens

2004 Olympic Games. International Journal of

Performance Analysis in Sport, 5(2), 1-8.

Performance Analysis in Volleyball: Problem of Merging Attack and Counterattack Spike as One Variable